phil ringnalda

Now that you’ve discovered that any bunch of flaming lunatics can knock a site out of Google if they have a lawyer, your second thought, after “everyone who voted for the DMCA should be drawn and quartered,” is probably “I need a backup search engine.” In the dark days before Google, I knew the urls of a dozen different search engines by heart, and did pages of forms to search multiple engines. Now, I can’t even remember whether Northern Lights is still around (it is, but you have to remember that you want Northern Lights Research to get to the web search).

(Note: if you are a random surfer looking for real search engine reviews, you should have picked searchenginewatch.com—these are mostly joking reviews intended to amuse me and possibly some friends.)

I did remember from the pre-Google days that a Norwegian company called Fast Search & Transfer had been doing good work. The search engine is now alltheweb.com. In an odd sort of way, their results are useful: the number one result is bloghop, and searching a second time there gets you to our target. Their beta “FAST categories” can be quite amusing: for my name they offer such categories as “Burningbird, Superior Knowledge”, “Pie Crust, Flaky, Site Exists”, and “Jonathon Delacour Unplugged, Mood Swings.” Points for humor, intended or not, and major points for being located outside the United States of Disney/TimeWarner, where the DMCA doesn’t apply.

I should have paid more attention to vivisimo.com than I have, since I’ve seen their odd referral, from vivisimo.com/Snippets.html, a time or two. They seem to be mostly a meta-search site, compiling results from other search engines (or, in the advanced search, your choice of news sites like YahooNews and the NY Times, or even auctions, the US government, or Britannica), adding value in their categories, and the Snippets, which is their alternative to Google’s cached site: you can get a live preview of the sites in a little frame right in the results, then open the page in the full window, in the results window, keeping the categories open in a frame at the left, or in a new window. The interface ends up looking horribly cluttered to Google-tuned eyes, but I suspect that once you learn your way around it, it’s vastly more efficient for searching out hard to find things, like very common phrases. Also, major points for relevance, but not for freshness, turning up an interview with proto-FolkYouHarder as the number one result.

Teoma.com has one of the best ideas, though not the best results sets, of the newer search engines. In theory, they return a list of web pages, the “web pages by topic” categories that everyone but Google seems to favor, and “experts’ links”, which are supposed to be links to pages that are authoritative lists of links on the search topic. You would think that the experts’ links, which they find programmatically, would be heavily populated with topical weblogs, but my search didn’t return any experts’ links, and only one category: “North American Breeders”. Again, points for humor, but not many points for quality results.

One interesting thing I did note while ego-searching all these sites: everyone else seems to be doing a much better job than Google of following links with queries in them. In theory, Google will follow one link with a ? in it, but then won’t follow any of the links on that page, to avoid clobbering servers with too many requests. In fact, I don’t see much evidence that they follow even a single query: phil ringnalda site:blogger.com, while it returns an interesting collection of fake subdomains (wrapyourass.blogger.com?), only returns the Discuss front page. The equivalent search at alltheweb returns 273 pages from the Discuss forum. I’ll have to see if I can figure out how to build a custom alltheweb search form that will limit to just blogger.com/discuss/ – it has to be better than the built-in search.